Microlithographic projection exposure methods are predominantly used nowadays for producing semiconductor components and other finely structured components, such as, for example, structured components for microsystems engineering. Highly integrated semiconductor components typically include a plurality of layers, of which only some layers are structured very finely, e.g. on the scale of a few dozen nanometers, while other layers have significantly coarser structures. The former layers realize in particular the actual main function of the semiconductor component, such as e.g. calculations and storage of data, while the latter layers serve e.g. for addressing and power supply. Structures having relatively coarse typical dimensions are also found in the field of microsystems engineering, such as e.g. in microelectromechanical systems (MEMS) or in microoptoelectromechanical systems (MOEMS). Semiconductor components are typically produced from a substrate of a semiconductor, while other substrate materials, in particular metals and vitreous substances, are also used in microsystems engineering.
Microlithographic projection exposure involves the use of masks (reticles) bearing the pattern of a structure to be imaged, e.g. a line pattern of a layer of a semiconductor component. A mask is positioned in a projection exposure apparatus between an illumination system and a projection lens in the region of the object plane of the projection lens and is illuminated with an illumination radiation provided by the illumination system. The radiation altered by the mask and the pattern passes as projection radiation through the projection lens, which images the pattern of the mask onto the substrate to be exposed. The substrate can be e.g. a semiconductor wafer. The substrate to be exposed bears a radiation-sensitive (i.e. photosensitive) layer composed of photoresist material on its side to be structured. The layer is also referred to as a resist layer.
In order that an image of the pattern that is as faithful to the original as possible is transferred to the substrate during the exposure process, the radiation-sensitive layer on the substrate surface should lie in the image-side focus region of the projection lens during the exposure time interval. In particular, the layer arranged on the substrate surface should lie in the region of the depth of focus (DOF) of the projection lens.
According to one common definition, the depth of focus specifies the distance relative to the plane of best focus for which the intensity of a point image is at least 80% of the intensity in the plane of best focus. This is equivalent to the condition that the diameter of the point image maximally doubles. The depth of focus amounts to half of the Rayleigh unit RU, which is defined as RU=λ/NA2, wherein λ is the operating wavelength of the projection exposure apparatus and NA is the image-side numerical aperture of the projection lens; the region in which the depth of focus condition is met accordingly has a total thickness equal to the Rayleigh unit RU. In general, the depth of focus becomes smaller, the higher the resolution capability of the projection lens.
Further miniaturization of the feature sizes on computer chips is becoming more and more difficult technically and physically. The associated costs are making further miniaturization of the feature sizes less and less attractive in some cases. As an alternative to better utilization of the wafer area, it is possible instead to utilize the third dimension by producing structures not just on or at the surface of the wafer but into the wafer in the depth direction.
The third dimension has e.g. already been used for the production of DRAM structures with the involved capacitors being etched into the depth (see e.g. US 2012/0049262 A1). Another trend is the three-dimensional stacking of flash memory structures (see e.g. U.S. Pat. No. 8,445,447 B2). Since the silicon layers grown are amorphous instead of crystalline, only relatively coarse structures can be produced with a functional capability for electronic reasons, which is why it is generally not necessary to use relatively short wavelengths such as 193 nm or even 13 nm for lithography processes of this type.
In order to be able to produce deep structures, holes having a large aspect ratio between depth and width are etched into the substrate. This can be done for example via a dry etching process, such as e.g. via ion etching. Etching methods of this type attack not only the substrate but also the developed and cured photoresist. The radiation-sensitive layer should therefore have a minimum thickness which is not too small.
For the exposure of relatively thick photoresist layers, in the field of 193 nm immersion lithography it is known to use so-called “Focus Drilling”. The ArF laser sources used in that case are so narrowband that it is not necessary for the corresponding projection lenses to be completely chromatically corrected. The wavelength dependence of the projection lens primarily consists in a (color-dependent) focus position dependent on the wavelength of the radiation used. By adjusting the line narrowing module (LNM) of an ArF laser in a targeted manner, it is possible for the bandwidth of the laser to be artificially increased. Different constituents of the spectrum which is continuous within the used bandwidth then simultaneously produce different focus positions. An effective increase in the depth of the region transilluminated within the photoresist layer thus occurs. However, this increase in the effectively usable depth of focus leads to blurring and thus to a reduction of contrast, which leads to coarser and/or less accurately definable structures on the finished structured substrate.
The article: “Thick Photoresist Imaging Using A Three Wavelength Exposure Stepper” by B. Todd, W. W. Flack and S. White in: SPIE MEMS 1999#3874-40, pages 1-15 describes the suitability of three ultrathick photoresists (layer thicknesses of 50 μm or 100 μm) for application in microstructuring with large aspect ratios using a wafer stepper having a mercury vapor lamp as radiation source. A catadioptric projection lens of the 1× Wynne-Dyson type having an image-side numerical aperture NA=0.16 is used for the imaging. Projection lenses of the Wynne-Dyson type allow a broadband exposure without introducing chromatic aberrations. For the exposure, the g-, h- and i-lines of mercury are used simultaneously in a wide spectral range of 350 nm to 450 nm (ghi-stepper). The results are compared with results of other experiments with a gh-line stepper and an i-line stepper of higher NA.
Another technology for producing holes extending deeply into a substrate with a high aspect ratio is multiple patterning using so-called hard masks. Examples are found in the article: “Evaluation of an advanced dual hard mask stack for high resolution pattern transfer” by J. Paul, M. Rudolph, S. Riedel, X. Thrun, S. Wege and C. Hohle in: Proc. of SPIE vol. 8685 86850V-1 to 86850V-11. The process is relatively complex and expensive. Moreover, each additional process step can lead to additional edge positioning errors of the structures produced on the substrate.